ABSTRACT Obesity is a top public health concern and its prevalence continues to increase in disadvantaged populations. Evidence suggests that obesity is associated with neighborhood built and social environment characteristics. These include walkability, exercise opportunities, green space, food outlets and social cohesion. But it is not clear whether these characteristics themselves cause obesity. Establishing causality is important to inform public health efforts to reduce health disparities. We propose to address this gap by studying a natural experiment in a high-poverty urban community. The city of Los Angeles is redeveloping Jordan Downs, an obsolete 700-unit public housing project. Over 5 years, the redevelopment will build?(1) 700 new units for existing residents, (2) 700 units for new mixed-income housing residents that will alter the social environment, (3) a new community center with a gymnasium, (4) retail space that will offer healthy eating opportunities, (5) green space that will increase opportunities for physical activity, and (6) additional built environment enhancements that will promote physical activity and healthy eating (e.g. walking and bike lanes, street lighting). Our goal is to study the effect of these improvements on Jordan Downs' residents' body mass index (BMI), overweight and obesity. In Aim 1, we will compare changes in these outcomes for the Jordan Downs residents and a control group (residents of 2 similar public housing projects) for 5 years. In Aim 2, we will disentangle the effects of three major components of the redevelopment: built environment, social environment and new housing. We will do this by taking advantage of the fact that the redevelopment will occur in several phases. In Secondary Aim 3, we will assess why, or why not, the redevelopment affected BMI, overweight and obesity. To do this, we will track physical activity, dietary behavior, and individual- and social mediators. Our data collection will include in-person surveys, 24-hr dietary recalls, and accelerometry. Finally, in Secondary Aim 4 we will explore the moderating effects of sex, race-ethnicity and age. Our study is time- sensitive because the first phase of redevelopment will be complete by Q4 of 2018. Thus, recruitment and baseline data collection must be complete by fall of 2018.